加了一些新功能

.gitignore

@@ -3,16 +3,20 @@
 /HZYSTempOutput
 
 #用于测试的输入输出
-*.wav
-textToRead.txt
+/*.wav
+/textToRead.txt
 
 #转换成可执行文件过程中生成的临时文件
 /excutables
 
 #打包好的文件
-*.exe
-*.bin
-*.zip
+/*.exe
+/*.bin
+/*.zip
 
 #用于测试模块的代码
-*.ipynb
+/*.ipynb
+
+#演示视频
+/*.flv
+/*.mp4

HZYS_GUI.py

@@ -1,8 +1,8 @@
 from platform import system
-from tkinter import Tk, Toplevel
+from tkinter import HORIZONTAL, Tk, Toplevel
 from tkinter import messagebox, filedialog
 from tkinter import Checkbutton, Button, scrolledtext, OptionMenu
-from tkinter import Label, font, BooleanVar, StringVar
+from tkinter import Label, font, BooleanVar, DoubleVar, Scale
 from huoZiYinShua import *
 from multiprocessing import Process, freeze_support
 from PIL import ImageTk, Image
@@ -36,7 +36,8 @@ def onDirectPlay():
 	textToRead = textArea.get(1.0, 'end')
 	myProcess = Process(target=HZYS.directPlay,
 						kwargs={"rawData": textToRead, "inYsddMode": inYsddMode.get(),
-								"pitchShift": optionConvertMap[pitchShiftOption.get()]})
+								"pitchShift": pitchShiftOption.get(), "norm": normAudio.get(),
+								"reverse": reverseAudio.get()})
 	myProcess.start()
 
 
@@ -49,7 +50,8 @@ def onExport():
 		if not outputFile.endswith(".wav"):
 			outputFile += ".wav"
 		HZYS.export(textToRead, filePath=outputFile, inYsddMode=inYsddMode.get(),
-					pitchShift=optionConvertMap[pitchShiftOption.get()])
+					pitchShift=pitchShiftOption.get(), norm=normAudio.get(),
+					reverse=reverseAudio.get())
 		messagebox.showinfo("疑似是成功了", "已导出到" + outputFile +"下")
 
 
@@ -154,13 +156,9 @@ def createConfigWindow():
 #储存生成选项的变量
 #-------------------------------------------
 inYsddMode = BooleanVar()
-pitchShiftOption = StringVar()
-optionConvertMap = {
-	"正常": "disabled",
-	"说的道莉": "xingzhuan",
-	"说的老理": "laotou",
-	"谁家小孩": "xiaohai"
-}
+normAudio = BooleanVar()
+reverseAudio = BooleanVar()
+pitchShiftOption = DoubleVar()
 
 
 
@@ -171,12 +169,6 @@ def createConfigWindow():
 									font=font.Font(family="微软雅黑", size=10))
 
 
-#复选框
-ysddCkBt = Checkbutton(mainWindow, text="匹配到特定文字时使用原声大碟",
-						variable=inYsddMode, onvalue=True, offvalue=False,
-						font=font.Font(family="微软雅黑", size=10))
-
-
 #按钮们
 #播放按钮
 playButton = Button(mainWindow, text="直接播放", command=onDirectPlay, height=1, width=8,
@@ -193,14 +185,33 @@ def createConfigWindow():
 					font=font.Font(family="微软雅黑", size=11))
 
 
+#原声大碟复选框
+ysddCkBt = Checkbutton(mainWindow, text="匹配到特定文字时使用原声大碟",
+						variable=inYsddMode, onvalue=True, offvalue=False,
+						font=font.Font(family="微软雅黑", size=10))
+
+
+#标准化音频复选框
+normCkBt = Checkbutton(mainWindow, text="分别标准化每个字和每条原声大碟句子",
+						variable=normAudio, onvalue=True, offvalue=False,
+						font=font.Font(family="微软雅黑", size=10))
+
+
+#倒放音频复选框
+reverseCkBt = Checkbutton(mainWindow, text="频音的成生放倒",
+						variable=reverseAudio, onvalue=True, offvalue=False,
+						font=font.Font(family="微软雅黑", size=10))
+
+
 #音调偏移文本
 pitchShiftLabel = Label(mainWindow, text="音调偏移选项：",
 					font=font.Font(family="微软雅黑", size=10))
 
 
 #音调偏移
-pitchShiftMenu = OptionMenu(mainWindow, pitchShiftOption,
-							*list(optionConvertMap.keys()))
+pitchShiftScale = Scale(mainWindow, from_=0.4, to=2.0, orient=HORIZONTAL, width=15, length=200,
+						resolution=0.1, variable=pitchShiftOption,
+						font=font.Font(family="微软雅黑", size=9))
 
 
 
@@ -218,16 +229,15 @@ def createConfigWindow():
 
 	#主窗口
 	#-----------------------------
-	#设置窗口大小
-	mainWindow.geometry("480x360")
+	mainWindow.geometry("480x420")
+	mainWindow.title("欧炫！纯纯的活字印刷")
+	mainWindow.resizable(False, False)
 	#窗口图标
 	try:
 		img = ImageTk.PhotoImage(Image.open("./lizi.ico"))
 		mainWindow.tk.call('wm', 'iconphoto', mainWindow._w, img)
 	except:
 		messagebox.showwarning("警告", "缺失图标")
-	#窗口标题
-	mainWindow.title("欧炫！纯纯的活字印刷")
 
 	#元素属性
 	#-----------------------------
@@ -238,9 +248,12 @@ def createConfigWindow():
 	configButton.place(x=310, y=230)
 
 	ysddCkBt.place(x=20, y=280)
-	pitchShiftLabel.place(x=20, y=310)
-	pitchShiftOption.set("正常")
-	pitchShiftMenu.place(x=110, y=310)
+	normCkBt.place(x=20, y=305)
+	reverseCkBt.place(x=20, y=330)
+
+	pitchShiftLabel.place(x=20, y=375)
+	pitchShiftOption.set(1)
+	pitchShiftScale.place(x=110, y=360)
 
 
 	#检查活字印刷实例是否配置正确

generateExeWin.bat

@@ -1,2 +1,2 @@
-CALL nuitka --mingw --disable-console --enable-plugin=numpy,tk-inter --nofollow-import-to=matplotlib --include-data-dir="E:/ProgramFiles/Python/Lib/site-packages/_soundfile_data"="_soundfile_data" --standalone --onefile --output-dir=./excutables/Windows/ ./HZYS_GUI.py
-CALL nuitka --mingw --enable-console --enable-plugin=numpy --nofollow-import-to=matplotlib --include-data-dir="E:/ProgramFiles/Python/Lib/site-packages/_soundfile_data"="_soundfile_data" --standalone --onefile --output-dir=./excutables/Windows/ ./HZYS.py
+CALL nuitka --mingw --disable-console --enable-plugin=numpy,tk-inter --nofollow-import-to=matplotlib --include-data-dir="E:/ProgramFiles/Python/Lib/site-packages/_soundfile_data"="_soundfile_data" --windows-icon-from-ico="./lizi.ico" --standalone --onefile --output-dir="./excutables/Windows/" "./HZYS_GUI.py"
+CALL nuitka --mingw --enable-console --enable-plugin=numpy --nofollow-import-to=matplotlib --include-data-dir="E:/ProgramFiles/Python/Lib/site-packages/_soundfile_data"="_soundfile_data" --windows-icon-from-ico="./lizi.ico" --standalone --onefile --output-dir="./excutables/Windows/" "./HZYS.py"

huoZiYinShua.py

@@ -19,13 +19,6 @@
 #目标采样率
 _targetSR = 44100
 
-#关键词与频率伸缩系数对照
-_keyWordPitchMap = {
-	"laotou": 0.8,
-	"xiaohai": 1.5,
-	"xingzhuan": 1.3
-}
-
 
 
 
@@ -40,21 +33,30 @@ def _fileName2FolderName(fileName):
 
 
 
+#标准化音频，统一音量
+def _normalizeAudio(data):
+	rms = np.sqrt(np.mean(data**2))
+	normData = data / rms * 0.2
+	return normData
+
+
+
 #读取音频文件
-def _loadAudio(fileDir):
+def _loadAudio(fileDir, norm):
 	data, sampleRate = sf.read(fileDir)
 	#双声道转单声道
 	if (len(data.shape) == 2):
 		#左右声道相加除以2
 		data = (data[:, 0] + data[:, 1]) / 2
-
 	#统一采样率
 	if (sampleRate != _targetSR):
 		#计算转换后的长度
 		newLength = int((_targetSR / sampleRate) * len(data))
 		#转换
 		data = np.interp(np.array(range(newLength)), np.linspace(0,newLength-1,len(data)), data)
-
+	#标准化
+	if norm:
+		data = _normalizeAudio(data)
 	return data
 
 
@@ -113,24 +115,24 @@ def config(self, configFileLoc):
 
 	#直接导出
 	def export(self, rawData, filePath="./Output.wav", inYsddMode=False,
-				pitchShift="disabled"):		
-		self.__concatenate(rawData, inYsddMode, pitchShift)
+				pitchShift=1, norm=False, reverse=False):		
+		self.__concatenate(rawData, inYsddMode, pitchShift, norm, reverse)
 		self.__export(filePath)
 		print("已导出到" + filePath + "下")
 
 
 
 	#直接播放
 	def directPlay(self, rawData, tempPath="./HZYSTempOutput/temp.wav",
-					inYsddMode=False, pitchShift="disabled"):
-		self.__concatenate(rawData, inYsddMode, pitchShift)
+					inYsddMode=False, pitchShift=1, norm=False, reverse=False):
+		self.__concatenate(rawData, inYsddMode, pitchShift, norm, reverse)
 		self.__export(tempPath)
 		playsound(tempPath)
 
 
 
 	#生成中间文件
-	def __concatenate(self, rawData, inYsddMode, pitchShift):
+	def __concatenate(self, rawData, inYsddMode, pitchShift, norm, reverse):
 		missingPinyin = []
 		self.__concatenated = np.array([])
 
@@ -190,7 +192,9 @@ def __concatenate(self, rawData, inYsddMode, pitchShift):
 						#拼接每一个字
 						try:
 							self.__concatenated = np.concatenate((self.__concatenated,
-																_loadAudio(self.__voicePath + word + ".wav")))
+																_loadAudio(self.__voicePath
+																			+ word + ".wav",
+																			norm)))
 						#如果出现错误
 						except Exception as e:
 							print(e)		#显示错误信息
@@ -208,7 +212,8 @@ def __concatenate(self, rawData, inYsddMode, pitchShift):
 					self.__concatenated = np.concatenate((self.__concatenated,
 														_loadAudio(self.__ysddPath
 																	+ self.__ysddTable[pronunciations[i][0]]
-																	+ ".wav")))
+																	+ ".wav",
+																	norm)))
 				#如果出现错误
 				except Exception as e:
 					print(e)		#显示错误信息
@@ -221,15 +226,17 @@ def __concatenate(self, rawData, inYsddMode, pitchShift):
 
 
 		#音高偏移
-		if (pitchShift != "disabled"):
-			self.__concatenated = _shiftPitch(self.__concatenated,
-											_keyWordPitchMap[pitchShift])
+		if (pitchShift != 1):
+			self.__concatenated = _shiftPitch(self.__concatenated, pitchShift)
 
+		#倒放
+		if(reverse):
+			self.__concatenated = np.flip(self.__concatenated)
 
 		#如果缺失拼音，则发出警告
 		if len(missingPinyin) != 0:
 			print("警告：缺失或未定义{}".format(missingPinyin))
-		
+
 
 
 	#导出wav文件